Semi-Autoregressive Transformer for Image Captioning

Yuanen Zhou, Yong Zhang, Zhenzhen Hu, Meng Wang; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2021, pp. 3139-3143

Current state-of-the-art image captioning models adopt autoregressive decoders, i.e. they generate each word by conditioning on previously generated words, which leads to heavy latency during inference. To tackle this issue, non-autoregressive image captioning models have recently been proposed to significantly accelerate the speed of inference by generating all words in parallel. However, these non-autoregressive models inevitably suffer from large generation quality degradation since they remove words dependence excessively. To make a better trade-off between speed and quality, we introduce a semi-autoregressive model for image captioning (dubbed as SATIC), which keeps the autoregressive property in global but generates words parallelly in local . Based on Transformer, there are only a few modifications needed to implement SATIC. Experimental results on the MSCOCO image captioning benchmark show that SATIC can achieve a good trade-off without bells and whistles. Code is available at \color magenta https://github.com/YuanEZhou/satic .